Evidence · first-party tested
The export flattens captions into the pixels instead of preserving them as a separate editable overlay layer, so post-export subtitle editing is not supported.
✗ FailedFlexClip →
A person speaking directly to the camera indoors with a static background, using natural hand gestures and facial expressions. This input is designed to test how well the tool handles a common talking-head editing scenario with stable framing and facial detail preservation.
The observation
The export flattens captions into the pixels instead of preserving them as a separate editable overlay layer, so post-export subtitle editing is not supported.
Criterion: Format support · Scenario: Indoor talking-head video (group: ai-video-background-removal)
Query this
get_evidence({
tool: "flexclip",
scenario: "ai-video-background-removal"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Free with attribution.
Same input, other tools
Real outputs, no retouching · every cell queryable via API & MCP · aidemos.com